More thoughts on AG-SG calibrations, drift assessment, and the transfer function of the iGrav system

We revisit a number of topics that frequently crop up in processing SG data. First is the question of the best procedure to use when calibrating the scale factor of the SG with AG measurements. One important decision is whether to use set data or drop data, as the statistics of the LSQ fitting routines are different in each case, depending on the fitting used, LSQ, L₁ norm etc. We show  examples using data from Apache Point and Strasbourg.. Secondly, the most beneficial method for assessing SG drift vs AG secular changes (e.g. discussed by Van Camp et al.) can be improved by taking into account local hydrology models at the site. Related to this we then suggest that where possible AG site measurements should, in addition to standard FG5-processing-derived numbers, include a value that includes corrections for standard environmental effects. Finally we provide pole-zero coefficients for the transfer function of the iGrav meter when using this instrument for tidal and seismic studies.